Abstract
OBJECTIVE: Despite decades of burnout research, clinical validated cut-off scores that discriminate between those who suffer from burnout and those who don't are still lacking. To establish such cut-off scores, the current study uses a newly developed questionnaire, the Burnout Assessment Tool (BAT) that consists of four subscales (exhaustion, mental distancing, and emotional and cognitive impairment). Separate cut-offs were computed for those at risk for burnout and those suffering from severe burnout for the original BAT-23 as well as for the shortened BAT-12. METHODS: Relative operating characteristic (ROC) analyses were carried out using representative samples of healthy employees from The Netherlands (N=1370), Belgium (Flanders; N=1403) and Finland (N=1350). In addition, samples of employees who received a burnout diagnosis were used (N=335, 158 and 50, respectively). RESULTS: The diagnostic accuracy of the BAT (area under the curve) ranges from good to excellent with the exception of mental distancing, which is fair. The country-specific cut-off values as well as their specificity and sensitivity are comparable to those of the pooled sample. CONCLUSIONS: In addition to country-specific cut-offs, general cut-offs can be used tentatively in other similar countries, pending future replication studies. Caution is warranted for using cut-offs for mental distance as the sensitivity and specificity of this subscale is relatively poor. It is concluded that the BAT can be used in organizational surveys for identifying employees at risk for burnout and, in clinical treatment settings, for identifying those with severe burnout, keeping in mind the tentativeness of the present cut-offs.